MiniMax M3.1 Flash 428B · Chip comparison
MiniMax M3.1 Flash 428B — B200 vs Vera Rubin NVL72
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and Vera Rubin NVL72 (NVIDIA Vera Rubin) on MiniMax M3.1 Flash 428B. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX →
At 440 tok/s/user interactivity on MiniMax M3.1 Flash 428B, B200 delivers 9,684 tok/s/chip at $0.050 per million tokens; Vera Rubin NVL72 delivers 160,104 tok/s/chip at $0.006. Vera Rubin NVL72 is 87% cheaper per token; Vera Rubin NVL72 delivers 17× the tok/s/chip at this point.
B200 posts 8,922 tok/s/chip for $0.054 per million tokens at 504 tok/s/user on MiniMax M3.1 Flash 428B; Vera Rubin NVL72 posts 87,831 tok/s/chip for $0.011. Vera Rubin NVL72 is 79% cheaper per token; Vera Rubin NVL72 delivers 9.8× the tok/s/chip.
Throughput at 569 tok/s/user on MiniMax M3.1 Flash 428B: B200 hits 8,231 tok/s/chip, Vera Rubin NVL72 hits 69,520. Costs per million tokens land at $0.058 and $0.014 respectively. Vera Rubin NVL72 is 75% cheaper per token; Vera Rubin NVL72 delivers 8.4× the tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B2009,684Vera Rubin NVL72160,104 | B2008,922Vera Rubin NVL7287,831 | B2008,231Vera Rubin NVL7269,520 |
| Cost ($/M tok) | B200$0.050Vera Rubin NVL72$0.006 | B200$0.054Vera Rubin NVL72$0.011 | B200$0.058Vera Rubin NVL72$0.014 |
| tok/s/MW | B2005,663,185Vera Rubin NVL7248,516,343 | B2005,217,418Vera Rubin NVL7226,615,499 | B2004,813,415Vera Rubin NVL7221,066,702 |
| Concurrency | B200~4Vera Rubin NVL72~213 | B200~4Vera Rubin NVL72~131 | B200~3Vera Rubin NVL72~85 |